Download Were Banks Special Intermediaries in the Late Nineteenth Century

M AY / J U N E 1 9 9 8
Eugene N. White is a professor of economics at Rutgers University. The author thanks Michael Bordo, Per Hansen, Naomi Lamoreaux, David
Weil, David Wheelock, and the other participants in the Federal Reserve Bank of St. Louis’ Twenty-Second Annual Economic Policy Conference
for comments.
Were Banks
Special Intermediaries in
Late Nineteenth
Century
America?
Eugene N. White
T
he financial crises and vastly increased
competition of the last two decades
have radically reshaped the American
financial system. One key feature of this
transformation has been the declining
importance of banks’ traditional activities.
Weakened by crises and regulatory disadvantages, banks’ share of intermediation
has shrunk while the shares of other financial intermediaries and markets have
expanded. The shrinking banking sector
has raised concerns because banks are
important “special” lenders to small firms
and other borrowers, they operate the payments system and provide liquidity, and
monetary policy is carried out by altering
their balance sheets. To put in historical
perspective the issue of banks’ declining
role in lending, this article examines the
nature of bank lending in the late nineteenth century and why banks remained
the dominant intermediaries, even when
disadvantaged by regulation and challenged by competitors.
In banking history, the late nineteenth
century is termed the National Banking
Era. Beginning in 1864 with the passage
of the National Banking Act and ending
with the founding of the Federal Reserve
System in 1913, the National Banking Era
was a period of rapid economic growth
and price stability. Growth was accompa-
nied by the spread of financial intermediation and innovation. Given the virtual
prohibition of branch banking and low
capital requirements, the demand for
banking services drove up the number of
commercial banks that were chartered
under the National Banking Act and state
laws from 467 in 1864 to 21,478 in 1913.
Commercial banks’ portfolios were shaped
by regulations that prohibited investment
in equities, limited mortgages, and encouraged short-term loans. Their liabilities
were predominantly demand deposits, and
although there had been experiments with
insurance of bank liabilities before the Civil
War, there was no insurance until very late
in the period, when seven states created
deposit guarantee funds after the Panic of
1907 (White 1983; Calomiris 1993).
As they do today, commercial banks
felt competitive pressures from other regulated financial markets and intermediaries,
including trust companies, investment
banks, insurance companies, and thrifts.
Combining deposit and loan banking with
other financial activities, trust companies
competed vigorously in the Northeast and
Midwest. Commercial banks could not
easily meet the demand for longer-term
finance by the newly emergent modern
corporations. Instead, investment banks
created the large bond and equity markets
to finance big business. These new financial
instruments were absorbed by life insurance
companies, often allied with investment
banks, which had a steadily rising flow of
policy premiums to invest. Banks also
faced competition from the money markets.
Improvements in transportation and communications enabled commercial paper
houses to intrude on banks’ territory,
offering access to a national market for
short-term credit. Mutual savings banks
catered to small depositors and the mortgage
market, although mortgage companies and
savings and loan associations became
increasingly important competitors late
in the century.
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
13
M AY / J U N E 1 9 9 8
Table 1
Financial Intermediaries’ Shares of Assets (percent)
1880
1900
1922
1950
1990
Commercial
Banks
Mutual
Savings Banks
Savings and Loan
Associations
Life Insurance
Companies
All Other
Intermediaries
62.6
62.9
63.2
50.8
27.0
22.6
15.1
8.8
7.6
2.1
1.3
3.1
3.7
5.7
8.9
10.5
10.7
11.6
21.1
11.1
3.1
8.2
12.7
14.8
50.9
F
SOURCES: Goldsmith (1958), U.S. Dept. of Commerce (1975), U.S. Comptroller of the Currency (1975), Snowden (1987),ederal
Reserve Bulletin (1991).
1
Banks’ share of intermediation
declined even if one takes a
broad definition of banking to
include mutual savings banks
and savings and loan associations. By this measure, banks’
share of intermediation falls
from 87 percent in 1880 or 81
percent in 1900 to 64 percent
by 1950 and 38 percent in
1990.
2
Noting that bank income from
off-balance-sheet activities rose
from 20 percent of total
income in 1979 to 33 percent
in 1991, Boyd and Gertler
(1993) argue banking has not
shrunk as much as would be
indicated by commercial banks’
share of assets. However, rising income from off-balancesheet activities is not new. In
the 1920s, this income rose
from 9 percent to 14 percent
of bank income (White 1984).
Unfortunately, there is not sufficient data to make long-term
comparisons of the relative
shares of intermediaries by
alternative measures.
In spite of these fast-growing
challengers, commercial banks retained
their preeminent position in the National
Banking Era. Table 1 reports the shares of
all financial intermediaries’ assets. Although
it is difficult to reconstruct a complete picture of the financial system before 1900,
the table demonstrates that commercial
banks retained their dominant position
among intermediaries well into the early
twentieth century. There was little change
between 1880 and 1922, when commercial
banks steadily held approximately 63 percent of assets. The twentieth-century
decline is evident in 1950; by 1990, commercial banks held only 27 percent of all
financial intermediaries’ assets.1 The
sources of this recent decline have been
studied intensively (Boyd and Gertler
1993; Wheelock 1993; and Berger,
Kashyap, and Scalise 1995). Banks’ commercial and industrial lending, a “special”
function of commercial banks, has been at
the center of this contraction. As a share
of all short-term debt of nonfinancial
corporations, banks’ commercial and
industrial loans fell from more than 80
percent in 1970 to 60 percent by the early
1990s (Wheelock 1993). Banks have lost
ground in lending to both nonbank intermediaries and markets. Finance company
loans have supplanted bank loans, while
offshore bank loans, not subject to reserve
requirements, have competed with domestic
banks and even domestic offices of foreign
banks to grab a bigger share of commercial
and industrial lending (Boyd and Gertler
1993). Instead of commercial and
industrial loans, many corporations
with good credit histories have found it
cheaper to borrow on the commercial
paper market.
However, banks’ traditional lending
operations have declined more than their
total operations (Berger, Kashyap, and
Scalise 1995). Banks have survived and
prospered by moving some traditional
business off their balance sheets. They
have unbundled traditional functions in
intermediation by offering loan commitments and standby letters of credit, and by
selling and securitizing loans.2 Banks
remain on the scene in the commercial
paper market by providing borrowers with
standby lines of credit. Boyd and Gertler
(1993) show that off-balance sheet items,
in terms of credit equivalents, are roughly
equal to half of banks’ commercial and
industrial lending. Berger, Kashyap, and
Scalise (1995) conclude that while banks
have lost considerable business to foreign
banks, nonbanks, and markets, their share
of intermediation has not shrunk as much
as measured by traditional activities on
their balance sheets.
Nevertheless, even with these qualifications, banks at the end of the twentieth
century are no longer the preeminent
financial institutions that they were at the
beginning. Recent theoretical work argues
that this smaller role for commercial banks
and depository institutions, in general, is a
predictable development. According to
Diamond (1997), banks will be key
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
14
M AY / J U N E 1 9 9 8
providers of liquidity and allocators of capital when there is limited participation in
markets in the early stages of economic
development. More liquidity is created by
banks’ offerings of demand deposits
backed by long-term assets, the price of
which is raised by the expansion of the
banking sector. The eventual growth of
markets increases the use of long-term
debt and equity. More participation in
markets induces banking sector shrinkage,
and bank holdings of long-term assets are
reduced relative to short-term assets. Diamond’s analysis emphasizes the role of
banks as providers of liquidity and
intermediation rather than as firms that
solve a problem of asymmetric information
by specializing in the evaluation and monitoring of high-risk, low-information
borrowers. The special informational
advantages of banks is, instead, the focus
of contemporary theoretical analysis of
banking lending, and the more central role
of commercial banks in the nineteenth
century suggests that these advantages
loomed even larger during the National
Banking Era. However, banking theory in
the nineteenth century was concerned
with very different issues and had very
strong prescriptions for lending that do
not accord with the modern literature.
To begin, this article reviews both contemporary and late nineteenth-century
banking theory. The limited available published data, complemented by two case
studies, provide some empirical evidence
on the special character of banking. The
preeminence of banking during the
National Banking Era highlights the key
attributes of commercial bank lending,
while the growing competition from markets and other intermediaries reveals the
origins of the change in banking’s relative
importance in the twentieth century.
deposits and making loans, seem less special to the operation of the financial
system. In recent theoretical literature,
banks are considered special not because
of the functions they perform but because
they overcome important informational
asymmetries.3 According to this literature
(Bhattacharya and Thakor 1993), the varied
forms of intermediation are responses to
various informational problems that
prevent markets managed by brokers from
efficiently selling borrowers’ liabilities to
savers. The special character of banks is
best understood by comparing them to
their closest market competition, money
market mutual funds (MMMFs). Both
banks and MMMFs provide transaction
services and increase divisibility and diversification for portfolios of large-denomination
assets. But banks are viewed by theorists as
different because they are delegated by
depositors to monitor borrowers. By monitoring borrowers through their transaction
activities and by covenant enforcement,
banks obtain continuous information on
their customers’ creditworthiness before
and after the creation of a new loan.
Banks invest in the acquisition of this
information to serve various types of borrowers. Bank loans may be useful to
borrowers who are relatively poor credit
risks and for whom information is relatively
volatile (Berlin and Mester 1992). Relatively
new borrowers, without well-established
reputations, may also gain more from bank
monitoring and choose bank loans instead
of the capital market (Diamond 1991).
Although small firms may find that banks
are some of the few sources of credit, the
higher interest they must pay for a bank
loan may indicate that the cost of such
loans is less than the cost of contracting
for marketable debt. For large firms with
access to the market, periodic signals from
short-term bank loans are a useful signal
to the market (Fama 1985).4 Although
banks do sell loans and securitize bundles
of relatively homogeneous collateralized
loans (automobile loans and mortgages),
loans are often difficult to market because
outsiders without access to banks’ proprietary information find it difficult to value
BANKS AS SPECIAL
INTERMEDIARIES IN
MODERN THEORY
The growing similarity among
financial intermediaries makes banks,
defined by their functions of taking
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
15
3
Fama (1985) first suggested
that banks had this special
character, examining why they
finance loans with both
demand deposits and certificates of deposit (CDs). Banks
are at a disadvantage vis-a-vis
other lenders because reserve
requirements are an implicit tax
on their liabilities. Banks can
compete by using demand
deposits, whose transaction services allow them to pay lower
interest. CDs do not provide
special transaction services and
must pay the same interest as
commercial paper and bankers’
acceptances. The viability of
CDs implies that borrowers
regard bank’s loans as special
and willingly pay a higher rate
of interest.
4
Some empirical evidence for the
special character of bank loans
has been found by James
(1987) and Lummer and
McConnell (1989), who discovered abnormal positive returns
on the stock of firms signing or
renegotiating credit agreements.
M AY / J U N E 1 9 9 8
5
A bill of exchange to finance
the shipment of goods was
thus an example of a real bill.
loans originated by banks. Many loans are
thus not marketed and, instead, are held
until maturity, with the banks bearing the
residual risk from nonperformance. It may
be difficult to sell loans without recourse,
because resale leaves the originating banks
with no incentive to produce the information needed to monitor the borrower.
Banks have a strong incentive to monitor
borrowers continuously because bank loans
are usually last in line of debt seniority.
Thus, renewal of a bank loan credibly signals other, more senior creditors of a firm
that they do not need to invest in a costly
analysis of the borrower.
Another key feature of banks that is
highlighted by this theoretical literature
is the fact that loans are financed with
shorter-maturity liabilities (Thakor 1992).
This maturity transformation requires
banks to bear interest-rate risk, for which
they are rewarded. A positive term premium
in the yield curve gives banks an incentive
to engage in a maturity mismatch. The
greater the mismatch, the higher the return
and volatility on a bank’s equity (Deshmukh,
Greenbaum, and Kanatas 1983). However,
a maturity mismatch also imposes market
discipline on the bank, which induces it to
screen and monitor loans, as deposits may
be withdrawn faster than loans are paid
off. This maturity mismatch makes banks
prone to panics in a system without
deposit insurance.
The threat of a panic arises because, as
delegated monitors, the banks themselves
need to be monitored. For transaction services, depositors need a very low-risk asset.
Demandable deposits, secured by a diversified portfolio of loans, are such an asset.
But security is guaranteed only when
depositors can discipline the bank
managers by quick redemption of their
deposits (Calomiris and Kahn 1991).
When depositors believe that their bank’s
risk has increased, they can withdraw their
deposits or refuse to roll over their shortmaturity CDs. However, given that loans
are difficult for outsiders to value, a change
in the economic environment may cause
depositors to panic. Some banks may
indeed be in trouble, but the inability to
value the portfolio of all banks correctly
may lead to contagion in which a run on
weak banks spreads to strong banks.
In the current literature, banks
perform a central role by lending on
nonmarketable information they have
produced and providing continuous monitoring of borrowers. Funded by liabilities
that create a maturity mismatch, banks are,
in turn, monitored by their depositors.
THE REAL-BILLS DOCTRINE
IN THEORY
Nineteenth-century banking theorists
would have found this contemporary
description of banks’ role in the financial
system somewhat puzzling. Banks were
regarded as very special intermediaries in
the nineteenth century, but for different
reasons than we think of today. There
was less competition from other intermediaries and markets, and hence there was
less concern about the special character
of bank loans. Instead, banking theorists
were more worried about the safety and
liquidity of loans as they affected the
banks’ ability to pay their depositors
on demand.
Virtually all students of banking in
the nineteenth century paid homage to
the “real bills” doctrine. According to
this theory, banks should offer only shortterm loans to finance the production or
shipment of goods. The sale of goods
would then be used to pay off the loans.
These loans were considered to be safe
because they financed real short-term
commercial transactions.5 Warning
against borrowing to pay off existing debts
or make speculative investments, Homans
(1857, p. 32) praised the safety of real
bills: “When money is to be invested in the
purchase of merchandise, cattle, flour, or
other property in the regular course of the
borrower’s business, the investment yields
to the borrower a means of repayment;
nothing is hazarded by ordinary integrity,
and ordinary exemption from disasters.”
Proponents of the real-bills doctrine
believed that if banks followed its prescriptions, the quantity of loans and liabilities
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
16
M AY / J U N E 1 9 9 8
problem. Bolles (1888, pp. 208-13)
was an exception, but his analysis was
somewhat strained, given real-bills
strictures. He admitted that mortgages
were “less readily convertible than some
other securities” but argued that “no property is more stable in value, and none less
likely to depreciate, than real estate.”
Bolles wrote, “These institutions represent
the industry and frugality of the masses,
and every effort should be made to put
them on the soundest footing.” The
best way to retain confidence was to
invest “savings deposits in mortgages
properly secured on the farms, the shops
and the homes of the people. If these are
not real values, what are?” Recognizing
that savings banks’ advantage probably
lay in local real estate, he warned against
lending on real estate out of the region
because of the difficulty in ascertaining
the mortgages’ underlying security. Like
other theorists of the period, he opposed
permitting large deposits in savings banks.
The organizers of mutual savings banks
had established rules to keep wealthy individuals from making deposits on the
grounds that these had been established
primarily to promote thrift among the
poorer classes. Bolles, on the other hand,
objected to participation by large depositors
because he believed that they would be
likely to withdraw their funds in a crisis.
Limit deposits to small savers and loans to
local real estate, said Bolles, and savings
banks would be safe institutions.
The real-bills prescription for lending
required that loans be short term. The
implication was that banks should minimize
the maturity mismatch of assets and liabilities. While this would reduce the earnings
that a bank might obtain from the term
premium as a result of a mismatch, it
would supposedly increase a bank’s ability
to meet a run on the bank, satisfying its
customers with the proceeds of loans that
were being rapidly paid off. Thus, even
though panics were frequent in the late
nineteenth century, depositors did not
play as large a disciplinary role in monitoring banks as envisioned by contemporary theory.
would be limited to the legitimate needs of
business, and banks would remain liquid.
In this era, liquid loans were simply loans
that were paid off at maturity. Liquidity
meant that an asset was automatically paid
off at maturity, not necessarily that it was
easy to market. The more modern idea of
liquidity and the idea of holding a diversified portfolio of readily marketable assets
for a secondary reserve did not gain wide
acceptance until the beginning of the Federal Reserve period (James 1978).
Proponents of the real-bills doctrine
offered blunt prescriptions to bankers.
In 1876, the Comptroller of the Currency
addressed the American Bankers’ Association: “As banks are commercial institutions, created for commercial purposes,
preferences in discounts should always be
given to paper based upon actual commercial transactions. Banks are not loan
offices. It is not part of their business to
furnish their customers with capital…”
(Bolles 1890, p. 70). The Comptroller
was emphatic that all paper should be
paid off at maturity, enabling banks to
meet withdrawals with funds from
maturing loans.
As late as 1915, Kniffin (1915, p.
209) wrote in a standard text on banking
that “the secret of sound banking is to
have a steady stream of money coming
in by way of maturing loans, so that the
constant stream of obligations falling due
daily by reason of the demands of the
checking depositors may be met. A
demand obligation cannot be met by a
time security and only as a bank keeps its
funds liquid—that is, flowing in and out—
can it meet every demand made on it
without hardship.”
The strict prescriptions that real-bills
advocates proferred to commercial banks
raise the question of how such advocates
would manage savings banks, which by
their very design held much longer-term
assets. Although funded by savings
deposits, savings banks did experience
runs in the late nineteenth century and
could not rely on a rapid payoff of loans
to meet their customers’ demands.6
Few writers of the period addressed this
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
17
6
Savings banks did have one
advantage in a crisis. While
commercial banks were required
to make payment on demand,
savings banks had the right
under state law to restrict payment under certain conditions.
M AY / J U N E 1 9 9 8
THE REAL-BILLS THEORY
IN PRACTICE
the character described, is based on
property of intrinsic value. … The
several makers of the paper, though
debtors in form, are only insurers, or
guarantors, in fact. They pledge their
respective property to the payment of
the loans; but the primary and generally sufficient pledge is the property
for which the notes are given. The
wealth of the makers is a necessary
margin or guaranty, because the property sold may be destroyed or the
value may fall.
While banking theorists offered
very strict prescriptions to banks in the
nineteenth century, banking practice
diverged from theory even as bankers
believed they held to the theory in spirit.
The following paragraphs explore some
of these discrepancies.
The Problem of Lending
The real-bills theoreticians favored the
use of two-name paper, but bankers began
to employ other financial instruments.
Before the Civil War, commercial transactions were usually financed by a trade
acceptance, a two-name bill of exchange
that provided recourse to the acceptor or
endorser of the bill in case of default. In
The Banker’s Common-Place Book (1857),
Homans advised his readers to accept only
notes endorsed by men of wealth and good
reputation and stated, “Banks… never regularly lend money without receiving the
security of more than one person who is
deemed safe for the debt; and a good
banker will err on the side of excessive
security, rather than accept security whose
sufficiency may reasonably be questioned.”
In his standard text on banking, Bolles
(1888, pp. 52-53) explained why twoname paper was essential for making
banking “a very safe and easy business”:
An ideal real bill, like a trade acceptance,
was secured by a real transaction, endorsed
by a respectable, wealthy individual, and
was short term. Evaluating the quality of
this form of lending relied as much on an
evaluation of the endorser as it did on the
issuer of the note and the safety of the
underlying transaction.
The crucial difference between
contemporary banking theory’s positive
description of banking and the real bills’
normative description is that they are
predicated on banks’ specializing in the
collection of different types of information.
According to contemporary theory, a
modern bank collects financial and transaction information from its customers to
assess their creditworthiness. In contrast,
nineteenth-century banks were told that in
addition to verifying that bills represented
bona fide transactions, they needed to
monitor and collect information on the
endorsers of bills. Bolles (1888, pp. 9799) described the process whereby a bank
would decide every day what loanable
funds were available and examine the bills
offered. Some makers and endorsers were
better known, and a bank would select the
most desirable offerings and decline the
remainder. In effect, the tasks of judging
the creditworthiness of the final borrower
and monitoring his performance were delegated partly to the endorsers of the bills,
to whom the bank had recourse.
In spite of the admonitions of realbills advocates, markets in the nineteenth
century moved away from two-name
paper. After the Civil War, the single-name
I should say that the first and most
important function of a bank is, by the
use of the capital which it controls, to
bridge over the periods of credit which
necessarily intervene between production and consumption, in such a
manner as to give back to each producer,
or middleman, as quickly as possible,
the capital invested by him in such
products, in order that he may use it
over again in new production or new
purchases. … Thus defined, banking
is not only one of the most useful; but
it is also one of the most safe and
healthy of business operations. Its
safety lies in the fact that every loan of
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
18
M AY / J U N E 1 9 9 8
Table 2
Composition of Bank Loan Portfolios
(percentage of total loans)
Total Loans
($ millions)
Demand,
Demand,
Unsecured
Secured
by Collateral by Collateral
Time,
Two-Name
Unsecured
Time,
One-Name
Unsecured
Time,
Secured by
Collateral
Mortgages or
Secured by
Mortgages
& Other
National
banks 1895
2,042
5.1
13.9
46.9
15.6
15.6
0
National
banks 1910
5,455
9.7
17.2
33.2
19.1
20.4
0.5
Nonnational
banks 1910
7,066
3.8
13.5
14.3
7.5
15.8
45.2
SOURCE: U.S. Comptroller
of the Currency, annual reports, 1895, 1910.
unsecured promissory note—commercial
paper—became the leading short-term
instrument for farmers and merchants.
This instrument was criticized for its lack
of adequate security in the form of collateral or personal guarantees from one or
more endorsers, and it required a more
modern direct evaluation and monitoring
of the customer. This increasingly popular
instrument could be taken directly by a
commercial bank or handled by a commercial paper house. Data on the types of
loans and their characteristics during the
National Banking Era are very scarce.
Greef (1938, p. 68) reports one estimate
that single-name paper constituted 75 percent of the market by 1894. Myers (1931,
pp. 322-23) calculated that the ratio of
two-name paper to total loans and
discounts fell from 50 percent in 1886 for
New York banks to 20 percent in 1900,
with the proportion of single-name paper
rising from 10 to 20 percent. For country
banks, these ratios fell from 50 percent to
33 percent for two-name paper and rose
from 10 percent to 20 percent for singlename paper. According to Myers (1931, p.
136), only 3 percent of all domestic credit
transactions were financed by trade acceptances at the end of the century. Some of
the limited data from the Comptroller of
the Currency’s annual reports is presented
in Table 2. For national banks, the share
of two-name unsecured paper fell from 47
percent to 33 percent of all loans and discounts between 1895 and 1910, while
unsecured time and demand loans both
rose. The amount of two-name paper held
by nonnational banks—all state banks,
savings banks, loan and trust companies,
and private banks—was already a low 14
percent by 1910.
Discounts were unsecured loans made
on the general credit of the borrower.
Loans were usually secured by a pledge of
collateral, including stocks, bonds, receivables, merchandise, or real estate. With
the decline in two-name paper, collateralized loans were of increasing importance
in the nineteenth century, as collateral provided an alternative to the guarantee of an
endorser. The Philadelphia National Bank,
for example, ventured into granting loans
on warehouse receipts, which proved to be
a profitable line of business.7 Considered
poor collateral for commercial banks
because it was illiquid, real estate did not
fit the real-bills doctrine, and mortgage
loans on real estate were prohibited to
national banks until 1913. The share of
both demand and time loans secured by
collateral in national banks’ loan portfolios
rose between 1895 and 1910 (see Table 2).
The largest item of secured lending for
national banks was, of course, mortgages.
Mortgage lending was dominated by
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
19
7
Although backed by collateral
involved in a real transaction,
as prescribed by real-bills theory, these loans were not completely safe. In 1888, the
bank found itself the owner of
a warehouse full of overvalued
prunes. Wainwright (1953),
p. 155.
M AY / J U N E 1 9 9 8
savings banks, and mortgages constituted
the largest item in their portfolios.
Given the hold of the real-bills
doctrine, with its emphasis on evaluating
specific transactions and the quality of
endorsers, there was a slow development
of alternative methods of evaluating loans
and the quality of a bank’s portfolio.
Bankers had supplemented their own
knowledge of business borrowers by subscribing to reports of credit agencies like
R.G. Dun and Company. Yet, these reports
were usually based on estimates of a firm’s
worth and reports from lawyers and other
business people about the character of its
proprietors, not financial statements (Lamoreaux 1994). Some banks required
borrowers to maintain compensating balances on deposit to gain additional
information.8 Contemporary writers also
urged banks to discover what constituted a
firm’s fixed assets and its quick or convertible assets, recommending that loans could
be granted if a borrower’s liabilities did not
exceed 50 percent of his quick assets. The
maximum recommended term was six
months. The result would be short-term,
self-liquidating loans.
Offering advice on how to judge a
potential borrower, Moulton (1918, p.
655) wrote the following:
8
Compensating balances could
be used to raise interest rates
above usury rates, but this tactic was less important in the
late nineteenth century, when
market rates had declined well
below usury rates in most
states. Lamoreaux (1994),
pp. 101.
The intimate and often qualitative
knowledge of local clients possessed by a
banker was not easily replaced by financial
statements in the late nineteenth century.
Perhaps the most important reason for the
failure to use financial statements was that
there were no uniform accounting
standards for business. This feature of
business practice added to the asymmetry
of information that gave banks their
special role. For all nineteenth-century
firms, accountants had little authority to
impose standardized accounting practices
on clients, and there were few statutory
requirements governing accounting
behavior (Brief 1966). Without uniform
accounting methods, there was no ready
alternative evaluation method to a banker’s
qualitative judgment of his customer.
The longstanding problem posed by an
absence of accounting standards came into
sharp focus when the Federal Reserve Board
wanted to guarantee the quality of member
banks’ paper eligible for discount by using
an objective analysis of financial condition
instead of subjective judgments about the
borrower’s character (Miranti 1986). To
ensure that lending officers analyzed reliable
data when granting credit, the Board wanted
borrowers’ statements to be certified by public
accountants, and it issued Circular No. 13
in 1914 to set down the rules. When notes
were offered for rediscount, they were to be
accompanied by a statement setting forth
the condition of the borrowers and stating
that the funds were used for the purpose of
financing current transactions, not fixed
capital or permanent working capital. The
member bank offering a note for rediscount
was required to have these statements on
file and to certify that they were in compliance with the circular (Willis 1923).
Member banks quickly realized that, under
these rules, they would be able to discount
very little of their paper. They protested to
the Fed that the average business, and
especially the average farmer, could not
furnish the appropriate type of statement,
the demand for which would be seen as
insulting and burdensome.
A special committee was convened by
the Board to reconsider Circular No. 13.
The amount that may be safely
loaned… can be ascertained only from
an intimate personal acquaintanceship
with the borrower and his business
or from a study of a balance sheet or
financial statement setting forth the
condition of the business. The
growing impersonality of modern
business in the larger centers and the
growing size and complexity of
business enterprise has more and more
necessitated the use of the balance
sheet as a basis of credit extension.
He believed that it was in the late 1870s
that financial statements were first used in
procuring loans, but it was not until the
1890s that their use became common even
in large banks. Few small suburban banks
or country banks used them.
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
20
M AY / J U N E 1 9 9 8
The committee offered the following
observation:
major financial center of Philadelphia, the
Philadelphia National Bank, had 34 men
on its payroll in 1879, including one assistant cashier to help manage its expanding
operations (Wainwright 1953). Management was in the hands of the directors, one
of whom was selected as president. The
directors verified the cashier’s accounts and
decided how much to lend and to whom
(Lamoreaux 1994). The directors thus had
no staff to draw up detailed reports on customers and instead relied on their local
knowledge of business and their customers’
“character.” The growth of business and
the shift to lending outside the community
created a need to professionalize bank management. Writers advocated that a professional bank staff should evaluate real bills
with objective standards, keeping banks
safe and sound and avoiding excesses from
insider lending. But most banks outside the
major urban centers were too small to be
able to follow these recommendations.
According to Margaret Myers, the first
credit department was established by the
Importer’s and Trader’s National of New
York in the 1880s (Myers 1931). At the
same time, the Philadelphia National Bank
found it necessary to add a credit department because directors no longer intimately
knew all borrowers (Wainwright 1953).
The idea began to spread slowly after the
Panic of 1893 (Westerfield 1924), but by
1899 there were still only 10 banks in New
York with credit departments. Credit
departments gradually made granting
credit more impersonal, examining the
financial records, not the character, of
prospective borrowers. However, their
role in the National Banking Era remained
small overall.
We believe that the country banks
which constitute the majority of our
members are generally without credit
files as known to the large city bank.
Borrowers are personally known by
the officers and directors who are usually their neighbors, and the means,
business and character of such
borrowers are matters of intimate personal knowledge to the bank officer. To
bring about a uniformunderstanding
among country bankers as to what is
and what is not eligible paper within a
narrow or even technically exact interpretation of the [Federal Reserve] Act
will take a long time and a still longer
time will be necessary to arrange for
the filing of financial statements by borrowing customers of country banks
(Willis 1923, p. 914).
The impossibility of imposing
financial and accounting standards on
member banks and their customers led the
Fed to back down and issue a new circular
in 1915 that lowered requirements for eligible paper. Most importantly, this circular
waived regulation for loans below $2,500,
which exempted most country bank paper.
The Federal Reserve would now discount
bank paper, but it was not easily marketable.
As this episode demonstrates, the absence
of generally accepted standard accounting
practices in the nineteenth century made
independent loan evaluation difficult, thus
augmenting the asymmetry of information
and ensuring the special position of banks,
which could observe lenders firsthand.
For banks, sophisticated borrower
evaluation—beyond reliance on personal
knowledge of local business—required
specialization. However, the banks of this
period were predominantly small with very
modest staffs and limited management
structures. Larger banks employed a
cashier who headed daily operations, several tellers and clerks, and perhaps a
bookkeeper. Smaller banks might have
only a cashier. The largest bank in the
The Maturity Mismatch
Another feature of lending practices
that contradicted the real-bills doctrine
was the length of loan contracts. The realbills doctrine assumed that banks would
operate without a large maturity mismatch
in order to reduce a bank’s exposure to a
run, but the common practice of renewing
loans produced a significant mismatch. If
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
21
M AY / J U N E 1 9 9 8
Table 3
Composition of Bank Deposits
(percentage of total loans)
Total Deposits
($ millions)
Demand
Deposits
Demand
Certificates
of Deposit
Time
Certificates
of Deposit
Savings
Deposits
Other
National
banks, 1910
5,287
80.1
7.6
8.2
0.0
4.1
Nonnational
banks, 1910
9,996
35.9
2.2
9.7
48.7
3.5
SOURCE: U.S. Comptroller of the Currency,
annual report, 1910.
one looked at bank portfolios without
inquiring into loan or borrower histories,
it appeared that banks did keep their
lending short-term. According to James
(1978), most bank loans had short-term
maturities. He concluded that typical
loans were for 30, 60, or 90 days, with one
year being an upper bound. An average
maturity was about 60 days. One survey
by the Comptroller of the Currency in
1913 found that 57 percent of all bank
loans had maturities of fewer than 90 days
(James 1978, p. 61). In 1913, the Comptroller of the Currency (1913, p. 100)
calculated that 57 percent of bank loans
had maturities of fewer than 90 days.
Lending practices differed quite sharply
from what appeared on banks’ books. Many
loans were rolled over in accordance with
the working-capital needs of firms and
farmers. Moulton (1918) saw little evidence that loans were automatically
liquidated at maturity. He found that
country banks granted repeated renewals,
extending a loan for years to finance
working capital. In commercial centers,
bankers estimated that 40 percent to 50
percent of unsecured loans were typically
renewed. The continuous needs for
working capital required continuous
credit. Unnerved by the Panic of 1907, the
Chicago banks asked Mr. Armour to liquidate his loans so they could replenish their
reserves. He replied, “What? I who am liquidating the country and taking the cattle,
sheep, and hogs that are being daily sent to
market to liquidate bank loans! . . . What
would be the condition of your bank loans
if I turned these cattle back to the farms?”
(Moulton 1918, pp. 719-20).
The maturity mismatch from funding
loans of a few months’ maturity with
demand deposits thus was even greater,
given the actual maturities of loans. If
commercial banks held substantial time
deposits, this mismatch would have been
reduced. However, commercial banks, and
especially national banks, primarily held
demand deposits. The reserve requirements set by the National Banking Act
of 1864 made no distinction among
demand deposits, savings, or time deposits,
and thus yielded no incentive to increase
deposits with longer maturities. Table 3
offers some limited data on the composition of bank deposits in 1910. Demand
deposits, at 80 percent of all deposits, were
of overwhelming importance for national
banks. The picture for nonnational banks
was more complex. States often set lower
reserve requirements for time deposits
(White 1983), encouraging the use of
these liabilities. One study of Minnesota
banks (James 1978) found that two-thirds
of deposits in country banks were time
deposits, while time deposits were just
10 percent to 25 percent of deposits in
city banks. Longer-term loans and a
maturity mismatch are what one would
expect to see, according to Diamond
(1997), when the capital market is not
yet well developed.
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
22
M AY / J U N E 1 9 9 8
The design of the Federal Reserve Act
had been partly informed by the real-bills
doctrine, and the Fed’s early regulations
reflected the theory. Even in the 1920s,
bankers were expected to conduct an
annual “clean up” of debt to demonstrate
their creditworthiness and to ensure that
the bank was not financing any permanent
capital. However, Jacoby and Saulnier
(1942) reported that while bankers
continued to offer short-term loans almost
exclusively, there was a full expectation on
the part of both borrowers and lenders that
these would be renewed. One study of
Iowa banks for 1914-24 showed that while
notes were dated with six-month
maturities, actual maturities ranged from
10 months to 32 months (Jacoby and
Saulnier, p. 13). Jacoby and Saulnier
observed that many businesses continued
to retire their loans for a short period each
year by borrowing from other institutions.
They commented that this had the limited
value of showing that the borrower could
get credit from another institution. Only
after the crisis of the 1930s did regulators
concerned about the absence of long-term
credit to business actively encourage
longer-term loans. Bank examiners were
instructed not to criticize loans because
they had maturities in excess of six
months, and the Banking Act of 1935 permitted Federal Reserve Banks to lend on
security of any sound asset, regardless of
maturity. Beginning in the late 1930s,
long-term loans, encouraged by federal
regulators and the cessation of new issues
on the capital markets, finally became
acceptable assets in bank portfolios, even
though long-term credits had been implicitly given in the nineteenth century.
The maturity mismatch may have
widened after the crisis of the Great
Depression and New Deal legislation.
Table 4 presents two surveys of member
bank loans in 1946 and 1955. One third
of member bank loans in these years had
maturities of more than one year. The
stated maturity structure of these banks
appears to be much longer than the structure claimed by most contemporaries and
historians for the National Banking Era.
Given that 75 percent of commercial bank
deposits were demand deposits in 1950
(Historical Statistics, Part II, p. 1022), this
would imply a greater maturity mismatch.
But this mismatch appears to have shrunk
in more recent years. The survey of new
loans for 1996 also presented in Table 4
shows that only 12 percent had maturities
over one year, and more than 72 percent
were for one month or less. This
shortening of maturities also appears to
have reduced the need for collateral, which
had fallen. By January 1996 (Federal
Reserve Bulletin, May 1996), demand
deposits accounted for only 15 percent of
all deposits. Even if NOW accounts are
included, the total is only 24 percent.
There is obviously close liability management and a closing of the maturity
mismatch. With the advent of highly
developed capital markets, commercial
banks, as Diamond (1997) argued, had
fewer long-term assets.
The other major nineteenth-century
depository institutions, the mutual savings
banks, had the bulk of their liabilities in
the form of savings deposits. In Table 3,
the large fraction of savings deposits in
nonnational banks reflects the inclusion of
mutual savings banks. Savings deposits
were assumed to be less volatile than
demand deposits and a good match for a
portfolio composed primarily of mortgages. Mutual savings banks’ exposure to
maturity mismatch was less than might
have been expected because mortgage contracts were different from today’s contracts.
A census study in 1895 of a sample of
mortgages from 35 counties across the
country found that the ratio of a mortgage to the underlying property value
was moderate, and the average life of a
loan was relatively brief. For farms
across regions, the ratio of the mortgage
to property value ranged from 32 percent
to 44 percent, and the average life of a
mortgage from 2.81 years to 6.62 years.
For homes, the ratio of values varied
between 33 percent and 48 percent, and
the loan life from 1.92 years to 5.99 years
(Snowden 1987). According to Snowden
(1991), these short mortgage contracts
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
23
Table 4
Terms of Lending for Commercial and Industrial Loans
Member Banks, November 20, 1946
Member Banks, October 5, 1955
Commercial Banks, February 5-9, 1996
Percent
Percent
collateralized
Billions
of dollars
Percent
Percent
collateralized
Billions
of dollars
Percent
Percent
collateralized
Total short-term loans
8.7
65.9
44.8
20.3
65.9
46.3
56.9
87.9
31.7
Demand loans
2.1
15.9
76.2
4.5
14.6
77.8
19.5
30.1
44.7
Overnight loans
13.6
21.0
11.8
1 month or less
13.8
21.3
21.5
9.9
15.3
44.7
24
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
Under 6 months
5.6
42.4
33.9
13.1
42.5
35.9
More than 1 month
6 months to 1 year
1.0
7.6
40.0
2.7
8.8
44.4
Total long-term loans
4.5
34.1
44.4
10.5
34.1
59.1
7.9
12.2
64.5
Total loans
13.2
100.0
43.9
30.8
100.0
51.0
64.7
100.0
31.7
tember 1959, and May 1996.
SOURCES: Federal Reserve Bulletin, June 1947, Sep
M AY / J U N E 1 9 9 8
Billions
of dollars
Type & Maturity of Loan
M AY / J U N E 1 9 9 8
permitted renegotiation to adjust to altered
circumstances. The relative brevity of
these contracts meant that the exposure to
risk from security mismatches was reduced
for any financial intermediary making
mortgages because a high proportion of
these mortgages fell due each year.
from their local banks. Most banks were
relatively small, and lending to a single
borrower was usually restricted to a
fraction of the bank’s capital (White
1983).10 These regulations helped to stimulate the development of the deep
American capital markets, where there
were no regulatory restrictions on the size
of an issue (White 1992). A firm requiring
a large short-term loan found commercial
paper an attractive alternative to borrowing
simultaneously from several banks.
Borrowers in the commercial paper
market were typically businesses with a
rapid turnover of merchandise or working
capital. A substantial net worth was
required for a firm to enter this market
(James 1996, pp. 222-23). While some
firms in this market relied on it exclusively,
most maintained lines of credit with commercial banks to meet usual short-term
credit needs. Kniffin (1915, p. 463)
advised that “it is good policy for a concern
to borrow in the open market and reserve
its home banks for emergencies. It can often
obtain better rates in the broad market, and
has the home bank to fall back upon when
needed.” The lengthening of credit terms
and a growth in receivables helped to spur
the development of single-name paper
(Baxter 1966, p. 5). Commercial paper had
maturities ranging between two months
and nine months, but most commonly
four months to six months. While banks
might feel obliged to renew loans, paper,
once granted, was paid off at maturity and
thus made a good investment for excess
funds. Initially both single-name and
double-name notes were issued in odd
amounts to mirror the exact credit demands
of the firms. But banks found this practice
inconvenient, and by 1890, commercial
paper came to be issued in common
denominations, usually $2,500, $5,000, and
$10,000 (Greef, pp. 75-77). By the 1890s,
not only merchants but also many manufacturers were active issuers in this market.
At the same time, commercial paper
houses became more professionalized.
They acted less as brokers between
borrowers and lenders than as outright
buyers, who held the paper for resale,
WERE LATE NINETEENTHCENTURY BANKS SPECIAL?
Competition from other intermediaries
and markets has recently called into question banks’ distinctive role in the financial
system. Gorton and Pennacchi (1993)
find some evidence for an unbundling of
banks’ two functions, making loans and
creating deposits. They posit that this
development is a result of technological
changes that have lowered the cost of
information production. Money market
mutual funds compete with demand
deposits by investing in commercial paper
instead of loans, while nonbank lenders,
including finance companies and revolving
credit, produce loans that compete with
banks’ commercial and industrial loans.
As these intermediaries do not tie demandable liabilities to nonmarketable assets,
there appears to be little threat of panic to
MMMFs from commercial paper defaults
or to nonbank lenders from the failure of
some of their number.9 As banks’ special
character is supposed to be embodied in
their ability to collect information and
monitor borrowers, the growth of MMMFs
seems to imply that the market has an
increased ability to make short-term credit
marketable. Although late nineteenth-century banks were different in many respects
from contemporary banks, having no rivals
like money market mutual funds, they
were forced to compete with the commercial paper markets which grew very fast in
this period.
The American commercial paper
market developed with the spread of the
unsecured promissory note (Greef 1938).
The structural defects of the banking
system, dominated by small unit banks,
spurred its growth on by allowing
businesses an alternative to borrowing
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
25
9
Gorton and Pennacchi caution
that these institutions may
mimic banks in that commercial
paper is backed by bank loan
commitments, and nonbank
lenders often finance their activities by issuing putable bonds.
10 Some
larger banks were adept
at innovation to provide more
credit. The Philadelphia
National Bank skirted around
the national bank rule limiting
loans to one borrower to 10
percent of capital by purchasing
railroad bonds, on which there
was no such limitation, with
the understanding that they
would be repurchased
(Wainwright 1953).
M AY / J U N E 1 9 9 8
11
The commercial paper market
shrank rapidly in the 1920s. In
1929, the ratio of open-market
paper to bank loans was 0.66.
It experienced a postwar revival,
becoming more important than
it had been in the pre-World
War I era. Commercial paper’s
share of all short-term lending
rose from 2 percent in 1966 to
15 percent in 1991 (James
1996, pp. 232-49).
and organized credit departments to evaluate borrowers. Whereas most banks had
relied on the recommendations of dealers
and correspondent banks, many began to
set up credit departments to investigate the
quality of notes.
Unlike today, commercial banks were
the largest purchasers of commercial paper
before the First World War. Although
commercial paper houses were the rivals of
banks for lending, banks bought most of
the paper issued. It was a useful alternative
investment for banks, paying a lower rate
of interest than loans they originated, but
it was safe, being carefully selected by the
houses that dealt in these obligations. By
1900, banks bought approximately 95 percent of all new offerings (James 1996).
The market was not very liquid. There
was no secondary market before the establishment of the Federal Reserve, although
some city banks might rediscount paper
for their correspondents. Commercial
paper was held to maturity when it was
paid off with near certainty. Competition
from commercial paper brokers vexed
bankers, especially rural bankers who lost
customers to commercial paper houses
(James 1978). Pressure from these bankers
led the American Bankers’ Association to
form a committee in 1908 to examine
competition from note brokers. The
committee attacked the bidding away
of good customers with low rates of
interest, lowering rates on commercial
loans below what they believed was
sustainable for banks.
Unfortunately, there is little information
on the size or growth of the commercial
paper market before the establishment of
the Federal Reserve System. After
examining the various estimates, Greef
(1938) concluded that just before the
founding of the Fed, total annual sales of
commercial paper was somewhat less than
$2 billion, representing obligations of
2,500 to 3,000 borrowing firms. James
(1978) accepts a figure of $1.7 billion for
1912.11 Assuming that commercial loans
and commercial paper had the same
average maturity, he computed the total
volume of new loans for banks in 1912
and found that commercial paper was 5
percent of total loans. Foulke’s (1931)
estimate of 5 percent to 12 percent of all
unsecured bank loans is in the same range.
Similarly, McAvoy (1922) estimated commercial paper to be 10 percent of loans
made by national banks.
In the first authoritative study of the
commercial paper market, Greef (1938)
found that borrowers used the commercial
paper market to obtain working capital
and for seasonal needs, the same reasons
that firms borrowed from banks. He noted
that most firms borrowed through
commercial paper dealers and from banks
at the same time or “rotated” their openmarket paper and bank loans. Even if
open-market rates were well below the
cost of bank loans, Greef found that firms
were careful to maintain satisfactory
average balances and open lines of credit
with banks. Greef observed that coordinated
borrowing from banks and the commercial
paper market offered firms advantages in
raising short-term capital. First, the cost
of open-market borrowing was usually
below the cost of bank loans, even after
adjustment for commissions to dealers and
retention of unused bank balances. Openmarket borrowing also gave firms bargaining
power with their banks and an ability to
“clean up” bank loans, when desirable,
and borrow larger sums than an individual
bank could supply. Contradicting modern
banking theory, Greef saw no major disadvantages to open-market borrowing
relative to bank borrowing, and he left
their coexistence largely unexplained.
In spite of the different character of
the commercial paper market in the nineteenth and early twentieth centuries, it
stood in a similar relationship to banks
then as it does today, and as posited by
contemporary banking theory. Firms
raised money in both markets, but they
appear to have resorted to banks for many
of the same reasons firms are alleged to do
so today. Firms needed to maintain open
lines to their banks, which they would use
presumably when they needed their creditworthiness verified. As Moulton (1918, p.
720) pointed out, commercial paper was
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
26
M AY / J U N E 1 9 9 8
not automatically renewable. If a firm
could not pay off a note when it was due,
it could look to its bank for credit instead
of the open market. The real-bills doctrine
claimed that firms should not be wholly
dependent on outside finance for workingcapital needs; hence the recommendation
for annual clean-ups. Although clean-ups
followed the precepts of real bills, they
also may be explained by modern theory—
and grudgingly seen by some contemporaries—as a useful way to subject the
firm to regular checkups to signal their
creditworthiness. Banks were thus
performing much the same function
as they do today.
banks are worth contrasting to the vast
majority of banks, which served as correspondent banks. While the large money-center
banks served as reserve banks for the rest
of the banking system, other banks concentrated more on commercial and industrial
loans to local business.
While the two case studies presented
here offer only a partial picture of bank
lending, they do show the unique role
of banks as lenders to small borrowers
who had limited access to other sources
of credit. In making and monitoring
its loans, the Bank of A. Levy, a commercial
bank, constantly observed the local farmers
and businessmen. The bank was familiar
with all aspects of local economic activity
and did not require financial balance sheets,
which could not have been produced in
any event. The Emigrant Savings Bank,
a mutual savings institution, carefully
mapped out the properties on which it
offered mortgages in New York and had a
great familiarity with its ethnic clientele
through its large base of savings deposits.
TWO CASE STUDIES
The paucity of aggregate quantitative
data on vital lending characteristics makes
it difficult to evaluate actual lending practices in the late nineteenth century. There
was a great deal of variation in banking
operations, depending on the region, the
location of the bank, and the size of its
operations. Under the National Banking
System, large city banks’ portfolios were
dominated by deposits they held as part of
other banks’ reserve requirements, which
they largely invested in brokers’ loans. In
Boston, Lamoreaux (1994) found that the
Merchants National Bank had 95 percent
of its loans backed by stocks and bonds,
and the Second National Bank had 79 percent. The one bank for which she found
detailed records, the Suffolk National
Bank, had 46 percent of its loan portfolio
in collateral loans (almost entirely brokers’
loans) and 54 percent in short-term loans
on personal security (mostly commercial
paper). One of the largest banks in New
York City, the National City Bank (Cleveland and Huertas 1985), handled the
financing of major corporations and entered
investment banking. Other prominent
banks, the First National Bank and National
Bank of Commerce, were allied with investment banks and life insurance companies
in the flotation of new securities (North
1954)—arrangements derided as the
“Money Trust.” These roles of the larger
The Bank of A. Levy
In 1885, the Bank of A. Levy began
operation as a small, rural, private
commercial bank in Ventura County,
California (White 1997). It grew rapidly
with the expansion of local agriculture
and took out a state charter in 1905,
surviving to become a prominent local
institution that was finally absorbed by
First Interstate in 1995. The surviving
loan book I examined covered the period
from August 1892 to October 1894 and
included 330 loans. Only 22 of the loans
had been purchased by the Bank, the rest
being made directly by the Bank. The
small percentage, 6.7 percent, of indirect
lending is similar to estimates for all
banks. These purchased loans do not
appear to have been obtained from a commercial paper house, and they were held
to maturity. The total value of all loans
was $124,120.
The loans made by A. Levy were
typical for a country bank, as described by
James (1978). The average loan was small,
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
27
M AY / J U N E 1 9 9 8
Table 5
Maturity of Bank of A. Levy Loans
Stated Maturity
(number of loans)
Percent
Actual Maturity
(number of loans)
Percent
283
85.8
0
0.0
2 to 30
6
1.8
25
7.9
31 to 60
6
1.8
34
10.7
61 to 90
5
1.5
29
9.1
91 to 120
7
2.1
28
8.8
121 to 365
19
5.8
128
40.4
365 +
4
1.2
73
23.0
Total
330
100.0
317
100.0
Days
1
under $400, but loans ranged from $5 to
more than $5,000. The loans were all
unsecured promissory notes. Achille
Levy’s bookkeeper wrote that Levy was a
disciple of the “character loan” method. If
he decided that an applicant was of good
character, a loan was forthcoming (Carroll
1958). If the borrower’s reputation was
flawed, the offer of thousands of dollars
worth of collateral could not persuade
Levy to make a loan. Levy carefully monitored his customers’ activities, not only
by observing their banking activities but
also by traveling around the county on
horseback, recording information in his
pocket notebook.
Although Levy’s lending violated
the real-bills doctrine by the use of singlename, unsecured promissory notes, it
followed the spirit, in that all the loans
were nominally short term. Almost 86
percent of all loans were one-day loans.
These loans would appear to have been
quite liquid, since Levy could call them for
repayment at a day’s notice. The typical
term of Levy’s borrowed funds for 1895
(the closest year with available data)
was one day—the same as the nominal
maturity of the loans. Thus, there appeared
to be no maturity mismatch with one-day
loans funded by demand deposits and
one-day bills payable.
However, the bank’s loans were
automatically rolled over. There is no
evidence that they were called before the
borrower was ready to repay the debt.
Table 5 shows the stated loan maturities
and the actual maturities for the 317 loans
for which there was information. The
average actual maturity for a loan was
279 days, or about nine months. Forty
percent lasted between four months and
one year, and 23 percent had actual maturities of more than one year. There was
only a small actual maturity mismatch
because the average actual maturity for
1895 bills payable was 199 days. The
actual loan maturities were at the high
end of the estimates for the length of
loans; few writers suggested that banks
make loans for more than one year. Levy
clearly provided a continuous source of
working capital for local farmers and merchants. Although there were 330 loans,
71 borrowers accounted for 70 percent of
the funds. Thirty-five borrowers obtained
credit twice in this two-year period, and
several individuals obtained anywhere
from five to eight loans. The pattern of
loan rates suggests that Levy monitored his
customers closely, lowering the rate once
they proved themselves by repaying the
first loan and raising the rate if they were
observed to borrow too heavily.
Was A. Levy’s bank special? In Ventura
County, several other banks competed for
customers. None of Levy’s customers
could have entered the national commercial
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
28
M AY / J U N E 1 9 9 8
paper market, although they might have
sold a promissory note to someone locally.
Achille Levy, the banker on horseback, was
very much like the banker of modern theory
whose role is to overcome the asymmetry
of information between borrower and
lender. His intimate knowledge of local
business—gained from his daily contacts,
monitoring of customer accounts, and frequent travels around the county—enabled
him to do a close evaluation of loan
prospects. He knew enough about his
clients that he could offer them unsecured
credit and have only seven out of 330
loans fail to pay him back in full. The fact
that most of these were one-day loans, for
which, in theory, full payment could have
been demanded the next day, may have
disciplined some borrowers. Levy did not
attempt to earn funds from a maturity mismatch. In principle, to the extent that his
depositors and creditors might run on
his bank, he could have liquidated his
loans very rapidly.
lected information on all the mortgages
made between 1866 and 1877. While the
depositor base remained firmly Irish and
Irish-American, loans were not as restricted.
They were made to individual home buyers,
to developers, and to religious organizations.
Loans to religious organizations accounted
for 7 percent of the total and 25 percent of
the value of loans made by EISB. The
developers are hard to identify, except
when they took out multiple loans on
adjacent plots. According to this method
of identification, 16 percent of the loans,
accounting for 10 percent of the value of
all loans, were given to developers. Most
loans were made for Manhattan property,
and the average loan size was $10,574 for
the 894 loans examined.
One striking feature of the EISB loans,
in contrast to modern mortgages, was the
lack of any provision for amortization
and the absence of any stated maturity
date. For most loans, borrowers simply
paid interest until they were able to pay
off the balance, although some made irregular payments on the principal. For
approximately the first dozen years, the
mortgage rate was set at 7 percent, the legal
maximum; but when interest rates declined,
the semiannual payments were reduced to
the new rate. Borrowers appear to have paid
off the loans when they saved up enough to
pay the principal. Table 6 shows the maturities for the 894 mortgages made over
11 years. While 29 percent of the mortgages
by number and 32 percent by value lasted
more than 20 years, a very large fraction had
maturities of under 10 years or even five
years. No one group of borrowers—residential, religious, or commercial and
industrial—stood out as taking shorter
or longer mortgages.
The EISB was a careful and prudent
investor, and there were very few defaults
on its mortgages. All properties were
local and appear to have been carefully
identified, recorded, and examined before
any loan was made. Mortgages were given
for only 50 percent of the value of the
property, offering the bank ample
protection. Bolles’ recommendations
for prudential savings-bank management
The Emigrant Savings Bank
The Emigrant Industrial Savings Bank
(EISB) was incorporated in 1850, thanks
largely to the efforts of John Hughes,
Catholic Bishop of New York. Hughes
prevailed on a group of 18 prominent
citizens, most of them Irish-born, to organize a safe deposit institution aimed at
encouraging thrift among poor Irish immigrants. The EISB was one of a score of
mutual savings banks set up in New York
before the Civil War and founded with
strong philanthropic motives. While these
banks did encourage the savings habit,
they also were operated on a sound
commercial basis (Olmstead 1976).
The EISB was one of the largest
mutual savings banks in New York City.
It was limited by its charter to invest in
bonds and mortgages worth double the
amount lent. About half its earning assets
were invested in mortgages. The mortgages
are recorded in the Bond/Mortgage Principal
and Interest books deposited in the New
York Public Library. To date, in a joint
study with Cormac O’Grada, we have col-
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
29
M AY / J U N E 1 9 9 8
Table 6
Maturity of EISB Mortgage Loans
Years
Number of loans
Percent
Dollar value (thousands)
Percent
0 to 4
183
20.5
1,795
23.4
5 to 9
214
23.9
2,814
36.7
10 to 14
155
17.3
1,481
19.3
15 to 19
85
9.5
921
12.0
20 and over
257
28.7
2,444
31.9
Total
894
100.0
7,660
100.0
appear to have been carefully followed.
Although the maturities for the 1895
census study were, on average, shorter
for the whole nation, the average for
the Northeast—six years—falls within the
modal range for the EISB. The EISB was
faced with a maturity mismatch because
it was funded with savings deposits that
were typically kept open only a few
years. However, the EISB invested a
substantial fraction of its portfolio in
call loans and bonds, which allowed
it to meet rapid decline in deposits in
periods of panic.
number of loans backed by collateral,
since lenders who knew less about their
borrowers needed some kind of protection.
Although banks did not have as many
near-competitors as they have today, they
did compete with the commercial paper
market in the creation of short-term credit.
Attempting to provide customers with sufficient capital, in spite of regulatory
restrictions and real-bills strictures, banks
offered considerable medium-term loans,
although their maturity periods were
shorter than those of most such loans in
the middle of the twentieth century.
The long twentieth-century decline
of commercial banks from their position
of preeminence has been told, partly, as
the result of regulatory disadvantage.
However, the decline may also be explained
as the consequence of technological
improvements, including established
accounting standards for business and
specialized management procedures for
assessing borrowers’ financial information.
The rise of credit analysis services and the
building of credit departments in banks
and commercial paper houses improved
banks’ ability to gauge the creditworthiness
of borrowers. These developments set the
stage for further improvements in information collection by other intermediaries
and markets. The dominant position of
commercial banks among financial intermediaries in the late nineteenth century
may thus be interpreted as the best
solution to the asymmetric information
problem between borrowers and lenders
when there were few technologically
CONCLUSIONS
Exemplifying the characteristics that
are believed to make banks “special” today,
banks in the late nineteenth century were
the dominant intermediaries. For small,
medium, and even many large borrowers,
banks were the only financial institutions
that offered credit. In the absence of wellestablished methods of accounting to
measure business performance, banks’
intimate knowledge of local business and
local business conditions was essential to
collecting information and monitoring
borrowers. The modern attributes of
banks as special lenders were formed principally during the post-Civil War shift
from two-name to single-name paper,
when banks began to concentrate on the
analysis and monitoring of the borrower
rather than the examination of the underlying transactions and endorsers of the
bills. This development also increased the
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
30
M AY / J U N E 1 9 9 8
_______. “Liquidity, Banks and Markets,” Journal of Political
Economy (October 1997), pp. 928-56.
feasible alternatives. The twentieth-century
decline in the prominence of banks as
intermediaries can be traced back to
the development of alternative markets
and the improvement of information
collection that began during the
National Banking Era.
Deshmukh, Sudhaker D., Stuart I. Greenbaum, and George Kanatas.
“Interest Rate Uncertainty and the Financial Intermediary’s Choice of
Exposure,” Journal of Finance (March 1983), pp. 141-47.
Fama, Eugene F. “What’s Different About Banks?” Journal of Monetary
Economics (January 1985), pp. 29-39.
REFERENCES
_______, and Michael Jensen. “Separation of Ownership and
Control,” Journal of Law and Economics (June 1983), pp. 301-25.
Agger, Eugene E. “The Commercial Paper Debate,” Journal of Political
Economy (July 1914), pp. 661-83.
_______, and _______. “Agency Problems and Residual
Claims,” Journal of Law and Economics (June 1983), pp. 327-49.
Baxter, Nevins. The Commercial Paper Market, Bankers Publishing
Company, 1966.
Foulke, Ray A. The Commercial Paper Market, Bankers Publishing
Co., 1931.
Berger, Allen N., Anil K. Kashyap, and Joseph M. Scalise. “The
Transformation of the U.S. Banking Industry: What a Long, Strange
Trip It’s Been,” Brookings Papers on Economic Activity (1995:2),
pp. 55-200.
Goldsmith, Raymond W. Financial Intermediaries in the American
Economy Since 1900, National Bureau of Economic Research and
Princeton University Press, 1958.
Berlin, Mitchell, and Loretta Mester. “Debt Covenants and Renegotiation,” Journal of Financial Intermediation (June 1992), pp. 95-133.
Gorton, Gary, and George Pennacchi. “Money Market Funds and
Finance Companies: Are They the Banks of the Future?” Structural
Change in Banking, Michael Klausner and Lawrence J. White, eds.,
Business One Irwin, 1993, pp. 173-214.
Bhattacharya, Sudipto, and Anjan Thakor. “Contemporary Banking
Theory,” Journal of Financial Intermediation (October 1993),
pp. 2-50.
Greef, Albert O. The Commercial Paper House in the United States,
Harvard University Press, 1938.
Board of Governors of the Federal Reserve System. Federal Reserve
Bulletin, Board of Governors of the Federal Reserve System, various
issues.
Homans, Issac Smith. The Bankers’ Common-Place Book, Bankers
Magazine, 1857.
Bolles, Albert S. Practical Banking, 5th and 7th eds., Homans Publishing
Company, 1888, 1890.
Jacoby, Neil H., and Raymond J. Saulnier. Term Lending to Business,
National Bureau of Economic Research, 1942.
Boyd, John J., and Mark Gertler. “U.S. Commercial Banking: Trends,
Cycles and Policy,” NBER Macroeconomics Annual 1993, Olivier Jean
Blanchard and Stanley Fisher, eds., MIT Press, 1993, pp. 319-68.
James, Christopher M. “Some Evidence on the Uniqueness of Bank Loans,”
Journal of Financial Economics (December 1987), pp. 217-35.
Brief, Richard P. “The Origin and Evolution of Nineteenth-Century Asset
Accounting,” Business History Review (Spring 1966), pp. 1-23.
James, John A. Money and Capital Markets in Postbellum America,
Princeton University Press, 1978.
Calomiris, Charles, and Charles Kahn. “The Role of Demandable Debt in
Structuring Optimal Banking Arrangements,” American Economic
Review (June 1991), pp. 497-513.
_______. “The Rise and Fall of the Commercial Paper Market,
1900-1929,” Anglo-American Financial Systems: Institutions and
Markets in the Twentieth Century, Michael D. Bordo and Richard Sylla,
eds., Irwin, 1995, pp. 219-59.
_______. “Regulation, Industrial Structure, and Instability in U.S.
Banking: An Historical Perspective,” Structural Change in Banking,
Michael Klausner and Lawrence J. White, eds., Business One Irwin,
1993, pp. 19-116.
Johnson, Alexander B. “A Treatise on Banking,” The Bankers’ CommonPlace Book, Issac Smith Homans, ed., Bankers Magazine, 1857.
Carroll, Russell. “The Bank of A. Levy,” Ventura County Historical
Society Quarterly IV (November 1958), pp. 10-13.
Kniffin, Jr., William H. The Practical Work of a Bank, Bankers Publishing
Company, 1915.
Cleveland, Harold van B., and Thomas F. Huertas. Citibank 1812-1970,
Harvard University Press, 1985.
Coffin, George M. The ABC of Banks and Banking, S.A. Nelson, 1901.
_______. The Savings Bank and Its Practical Work, 3rd ed., Bankers
Publishing Company, 1918.
Diamond, Douglas W. “Monitoring and Reputation: The Choice between
Bank Loans and Directly Placed Debt,” Journal of Political Economy
(August 1991), pp. 689-721.
Lamoreaux, Naomi. Insider Lending: Banks, Personal Connections, and
Economic Development in Industrial New England, Cambridge
University Press and National Bureau of Economic Research, 1994.
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
31
M AY / J U N E 1 9 9 8
Lummer, Scott L., and John J. McConnell. “Further Evidence on the
Bank Lending Process and the Capital-Market Response to Bank Loan
Agreements,” Journal of Financial Economics (November 1989),
pp. 99-122.
_______. “California Banking in the Nineteenth Century: The Art and
Method of the Bank of A. Levy,” unpublished manuscript.
Miranti, Jr., Paul J. “Associationalism, Statism, and Professional
Regulation: Public Accountants and the Reform of the Financial
Markets, 1896-1940,” Business History Review (Autumn 1986),
pp. 438-68.
U.S. Department of Commerce. Historical Statistics of the United States:
Colonial Times to 1970, U.S. Bureau of the Census, 1975.
U.S. Comptroller of the Currency. Annual Reports, Government Printing
Office, various years.
McAvoy, Walter. “The Economic Importance of the Commercial Paper
House,” Journal of Political Economy (February 1922), pp. 78-87.
Moulton, H.G. “Commercial Banking and Capital Formation,” Journal of
Political Economy (May, June, July, November 1918), pp. 484-508,
638-63, 705-31, 849-81.
Myers, Margaret G. The New York Money Market, Volume I: Origins and
Development, Columbia University Press, 1931.
North, Douglass C. “Life Insurance and Investment Banking at the
Time of the Armstrong Investigation,” Journal of Economic History
(Summer 1954), pp. 209-28.
Olmstead, Alan L. New York City Mutual Savings Banks, 1819-1861,
University of North Carolina Press, 1976.
Snowden, Kenneth A. “Mortgage Rates and American Capital Market
Development in the Late Nineteenth Century,” Journal of Economic
History (September 1987), pp. 671-92.
_______, and Walid Bu-Saba. “Mortgage Loan Duration and
Renegotiation in the Late Nineteenth Century,” unpublished manuscript.
Steiner, W.H. “The Commercial Paper Business,” Federal Reserve
Bulletin (August-September 1921), pp. 920-26, 1052-57.
Thakor, Anjan V. “Maturity Transformation,” The New Palgrave
Dictionary of Money and Finance, Vol. 2, London: Macmillan, 1992,
pp. 678-80.
Wainwright, Nicholas B. History of the Philadelphia National Bank;
A Century and a Half of Philadelphia Banking,1803-1953, William F.
Fell Co., 1953.
Westerfield, Ray B. Banking Principles and Practice, Ronald Press Co., 1924.
Willis, Henry Parker. The Federal Reserve System: Legislation,
Organization and Operation, Ronald Press Co., 1923.
Wheelock, David C. “Is the Banking Industry in Decline? Recent Trends
and Future Prospects from a Historical Perspective,” this Review
(September/October 1993), pp. 3-22.
White, Eugene N. The Regulation and Reform of the American Banking
System, 1900-1929, Princeton University Press, 1983.
_______. “The Effects of Bank Regulation on the Financing of
American Business, 1860-1960,” Finance and the Enterprise, Vera
Zamagni, ed., Academic Press, 1992.
F E D E R A L R E S E R V E B A N K O F S T. L O U I S
32